From mboxrd@z Thu Jan 1 00:00:00 1970 Received: from mailman by lists.gnu.org with tmda-scanned (Exim 4.43) id 1MPNlm-0007br-UH for qemu-devel@nongnu.org; Fri, 10 Jul 2009 17:36:38 -0400 Received: from exim by lists.gnu.org with spam-scanned (Exim 4.43) id 1MPNlh-0007Y2-Gd for qemu-devel@nongnu.org; Fri, 10 Jul 2009 17:36:37 -0400 Received: from [199.232.76.173] (port=36666 helo=monty-python.gnu.org) by lists.gnu.org with esmtp (Exim 4.43) id 1MPNlh-0007Xz-AF for qemu-devel@nongnu.org; Fri, 10 Jul 2009 17:36:33 -0400 Received: from an-out-0708.google.com ([209.85.132.241]:13737) by monty-python.gnu.org with esmtp (Exim 4.60) (envelope-from ) id 1MPNlg-0002mP-ML for qemu-devel@nongnu.org; Fri, 10 Jul 2009 17:36:32 -0400 Received: by an-out-0708.google.com with SMTP id c38so593357ana.37 for ; Fri, 10 Jul 2009 14:36:31 -0700 (PDT) MIME-Version: 1.0 In-Reply-To: References: Date: Fri, 10 Jul 2009 14:36:31 -0700 Message-ID: <2a50f7880907101436w699de79bsc7603a9bb4aae451@mail.gmail.com> Subject: Re: [Qemu-devel] 6a7ad29 regression From: Jordan Justen Content-Type: multipart/alternative; boundary=0016e64548c007467f046e60c424 List-Id: qemu-devel.nongnu.org List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , To: Consul , qemu-devel@nongnu.org --0016e64548c007467f046e60c424 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit I found that 6a7ad2998ca47711eddd191caa1a95f14459793f also seems to break the automatic fat disk generation functionality. (For example, '-hdb fat:path/to/make/into/fat/disk') I reverted 6a7ad29 on top of 52249f0f3c9f101a71ed24e207731525b8431c0a, and it appeared to fix the issue. -Jordan On Wed, Jun 17, 2009 at 5:22 PM, Consul wrote: > This commit (6a7ad2998ca47711eddd191caa1a95f14459793f) makes Qemu freeze at > startup (before OS is loaded). > Windows XP host and guest, qcow2 image. > Reverting it on top of the most current (42f1ced22) commit makes things run > again. > > qemu -L . -hda c:\qemu-img\xp2.q2 -boot c -localtime -soundhw es1370 -m 512 > -net nic,model=rtl8139 -net tap,ifname=TAP0 -kernel-kqemu > > > Alex. > ------- > > > Author: Dor Laor 2009-06-01 02:07:23 > Committer: Anthony Liguori 2009-06-16 13:36:47 > Tags: freezes > Parent: 3fe5c14e7472f88dddf837a9faa60f6abb2ec6bf (Remove dead code) > Child: 8c14c17395809dfb2b20bd3598e067d2c5af7dc4 (kvm: Fix IRQ injection > into full queue) > Branches: master, pcap2, remotes/origin/master, test > Follows: bad > Precedes: > > Call qemu_bh_delete at bdrv_aio_bh_cb. > > Also replave qemu_bh_cancel with qemu_bh_delete in bdrv_aio_cancel_em. > Otherwise the bh will live forever in the bh list. > > Signed-off-by: Dor Laor > Signed-off-by: Anthony Liguori > > > > --0016e64548c007467f046e60c424 Content-Type: text/html; charset=ISO-8859-1 Content-Transfer-Encoding: quoted-printable
I found that 6a7ad2998ca47711eddd191caa1a95f14459793f also seems to bre= ak the automatic fat disk generation functionality.=A0 (For example, '-= hdb fat:path/to/make/into/fat/disk')

I reverted 6a7ad29 on top o= f 52249f0f3c9f101a71ed24e207731525b8431c0a, and it appeared to fix the issu= e.

-Jordan


On Wed, Jun 17, 2009 at 5= :22 PM, Consul <v= oid@aleksoft.net> wrote:
This commit (6a7ad2998ca47711eddd191caa1a95f14459793f) makes Qemu freeze at= startup (before OS is loaded).
Windows XP host and guest, qcow2 image.
Reverting it on top of the most current (42f1ced22) commit makes things run= again.

qemu -L . -hda c:\qemu-img\xp2.q2 -boot c -localtime -soundhw es1370 -m 512= -net nic,model=3Drtl8139 -net tap,ifname=3DTAP0 -kernel-kqemu


Alex.
-------


Author: Dor Laor <do= r@redhat.com> =A02009-06-01 02:07:23
Committer: Anthony Liguori <aliguori@us.ibm.com> =A02009-06-16 13:36:47
Tags: freezes
Parent: 3fe5c14e7472f88dddf837a9faa60f6abb2ec6bf (Remove dead code)
Child: =A08c14c17395809dfb2b20bd3598e067d2c5af7dc4 (kvm: Fix IRQ injection = into full queue)
Branches: master, pcap2, remotes/origin/master, test
Follows: bad
Precedes:

=A0 =A0Call qemu_bh_delete at bdrv_aio_bh_cb.

=A0 =A0Also replave qemu_bh_cancel with qemu_bh_delete in bdrv_aio_cancel_= em.
=A0 =A0 Otherwise the bh will live forever in the bh list.

=A0 =A0Signed-off-by: Dor Laor <dor@redhat.com>
=A0 =A0Signed-off-by: Anthony Liguori <aliguori@us.ibm.com>




--0016e64548c007467f046e60c424--